1832.] Analysis of the Purdnas. 543 



from the Bhdgavat, but is shorter and less particular. There is also 

 this rather unintelligible addition, that the stars of the sphere never set ; 

 but the passage may signify, that they are not annihilated at the usual 

 periods of destruction. The text is in this place evidently incorrect, 

 and the translation being made from a single copy, it is not safe to 

 venture any emendation. 



A legendary account of Nildkanthd or (he blue-necked Slva follows, 

 and the description of the classes of the Pitris, and their feeding upon 

 the lunar nectar ensues. The introduction of obsequial ceremonies and 

 the worship of the manes appears to have originated with Pururavas, a 

 not unlikely circumstance, and one which explains the legend of his be- 

 ing descended both from the sun and moon ; the worship of the manes 

 being connected with the conjunction of these luminaries. The list of 

 Pitris differs in some respect from that of Menu, and from that given in 

 a manual used by the Brahmans of Bengal, in which a verse cited from 

 the Vdyu Purdna enumerates the following as the seven classes, Saumy- 

 as, Agniswattas, Vdrhishadas, Havishmantas, Ushmapas, and Ajy- 

 apas. In the chapter now under consideration there are but four parti- 

 cularised. The Saumyas or Somapas ; the Kavyas or Ajyapas ; the 

 Verhishadas, and Agniswattas. Three others are merely named, the 

 Ushmapas, Devakirttyas, and apparently the Lekhas and Bahwikasy- 

 as ; but these are unusual and probably inaccurate appellations. The 

 whole of the section is obscure, incorrect, and often unintelligible. The 

 same may be said of the two remaining chapters, which treat of the divi- 

 sions of time and the influence of the four ages. 



Without being in possession of the contents of the remaining portion, 

 at least one-half of the Vdyu Purdna, it is impossible to offer any opini- 

 on on the date as derivable from internal testimony. As far as the por- 

 tion analysed extends, it may be considered perhaps, as the oldest of the 

 actually existingPurdnas, and it has every appearance of being a genuine 

 work, conforming more closely than any yet examined to the defi- 

 nition of a Purdna, and admitting few of the unconnected digressions 

 and legendary absurdities by which the course of these compilations is 

 so commonly interrupted, and the established order widely disarranged 

 or wholly obliterated. 



The Vdyu Purdna is not unfrequently omitted in lists of the eighteen 

 Purdnas, but in that case it is considered to be the same with the 

 Saiva Purdna, which takes its place. As now met with, however, the 

 two works are not identical. 



2 x 



